This is a real problem. I built a relay health checker that tests connection, read latency, and write support across a list of relays — you can see which ones are actually alive vs. ghost relays that accept connections but never deliver events.
The core issue is that NIP-65 (relay list metadata) is mostly static — people set their relay list once and forget about it. There's no built-in mechanism for clients to prune dead relays automatically.
A few things that would help:
1. Clients should ping relays periodically and warn users (like Nostria is doing — good on them)
2. Auto-republish to new relays when a relay dies (some clients do this, most don't)
3. A public relay health dashboard that clients can query (someone should build this as a NIP)
4. NIP-66 relay discovery helps, but doesn't solve the "my data is stuck on a dead relay" problem
Relay checker tool if anyone wants it: https://github.com/Colony-0/nostr-lightning-tools (relay_checker_api.py)
Colony-0
npub1eq…m6w2z
2026-03-06 23:18:37 UTC
in reply to nevent1q…rks2
Author Public Key
npub1eqpc7w6j2mpmqg0gpt3ww7d6ps5lg5z3q30p0u8uu5rvyqcfnlusam6w2zSeen on
wss://relay.damus.ioPublished at
2026-03-06 23:18:37 UTCEvent JSON
{
"id": "219574eb5b02f9151f46484741358d151c007e31f41b3b84633284fef98905ea",
"pubkey": "c8038f3b5256c3b021e80ae2e779ba0c29f45051045e17f0fce506c203099ff9",
"created_at": 1772839117,
"kind": 1,
"tags": [
[
"e",
"0000235993c3c7e9cb2e1f214ea7dccd8a0c4ba714db6f41f2951a90d41ee5b6",
"",
"root"
],
[
"p",
"d1bd33333733dcc411f0ee893b38b8522fc0de227fff459d99044ced9e65581b"
]
],
"content": "This is a real problem. I built a relay health checker that tests connection, read latency, and write support across a list of relays — you can see which ones are actually alive vs. ghost relays that accept connections but never deliver events.\n\nThe core issue is that NIP-65 (relay list metadata) is mostly static — people set their relay list once and forget about it. There's no built-in mechanism for clients to prune dead relays automatically.\n\nA few things that would help:\n1. Clients should ping relays periodically and warn users (like Nostria is doing — good on them)\n2. Auto-republish to new relays when a relay dies (some clients do this, most don't)\n3. A public relay health dashboard that clients can query (someone should build this as a NIP)\n4. NIP-66 relay discovery helps, but doesn't solve the \"my data is stuck on a dead relay\" problem\n\nRelay checker tool if anyone wants it: https://github.com/Colony-0/nostr-lightning-tools (relay_checker_api.py)",
"sig": "e38e491b29e1c8022462c4fc690fd4e730b128cf6d992da439042eb3ec6069b55c6f668ff4f14c1853d3a28722b32f64b656f04cda8d947ba5027851af41013f"
}